Disparities in cancer treatment and outcomes have been extensively documented by the race, ethnicity, and socioeconomic status (SES) of the patient. Patient navigation represents a potentially powerful intervention for addressing these disparities, but has not been rigorously evaluated. The primary aim of this project is to assess the impact of a primary care-based, patient navigation-activation program on cancer-related quality of care using a randomized controlled study design. A key secondary aim assesses whether members of underserved populations derive greater benefit from navigation than do patients who face fewer access barriers. Six, large inner-city practices serving predominately minority and low-income populations will be recruited. Implementation of the intervention will be preceded by evaluation of the practices' needs, assessment of patient barriers, and development of a navigator curriculum and training manuals. Patients from these practices will be recruited using patient registries established in each practice for breast and colorectal cancer screening. Consenting patients with abnormal breast or colorectal screening results will be randomized to usual care or patient navigation. Well-trained community health workers, will assess patient barriers, assist patients in overcoming these barriers, and coach patients in communicating with their physicians. The primary patient outcomes will be time to a definitive diagnostic test and time to completion of cancer treatment, adherence to recommendations, receipt of guideline concordant care, satisfaction, and quality of life. Secondary outcomes include differential effects by race/ethnicity, income, or health literacy, mediating effects of patient activation, and costs of navigation and care. Findings from this project have implications for national policy related to the design and implementation of patient navigation programs for cancer and interventions designed to eliminate disparities in cancer treatment.